Skip to main content

​
Overview

MedMitra’s case management system provides a centralized platform for healthcare providers to create, organize, and track patient cases. Each case contains patient information, medical documents, and AI-generated insights all in one place.

​
Key Features

Create Cases

Quickly create new patient cases with demographic information and initial assessments

Document Upload

Attach lab reports and radiology images directly to patient cases

Case Tracking

Monitor case status from creation through analysis completion

Retrieve Insights

Access AI-generated insights, SOAP notes, and diagnostic recommendations

​
Creating a New Case

Cases are created through the /cases/create_case endpoint with the following information:

​
Required Information

  • Patient Name: Full name of the patient
  • Patient Age: Age in years
  • Patient Gender: Gender identity
  • User ID: Healthcare provider’s identifier

​
Optional Information

  • Case Summary: Initial clinical notes or chief complaint
  • Lab Files: Laboratory reports (PDF format)
  • Radiology Files: Medical imaging files

​
Example Workflow

# Case creation with file uploads
POST /cases/create_case

Form Data:
- user_id: "b8acad4b-4944-4d66-b405-de70886e7248"
- patient_name: "Jane Daniel"
- patient_age: 31
- patient_gender: "Female"
- case_summary: "Patient presents with respiratory symptoms"
- lab_files: [blood_test.pdf, urinalysis.pdf]
- radiology_files: [chest_xray.jpg]
The system automatically:
  1. Generates a unique case ID (UUID)
  2. Uploads files to secure storage
  3. Initiates background AI analysis
  4. Returns case details with file references
Source: backend/routes/case.py:43-122

​
Retrieving Cases

​
Get All Cases

Retrieve all cases for a specific healthcare provider: 
GET /cases/all_cases?user_id={user_id}
Returns a list of all cases with basic information and status. Source: backend/routes/case.py:131-143

​
Get Specific Case

Retrieve detailed information about a specific case: 
GET /cases/cases/{case_id}
Returns:
  • Complete case details
  • All attached files with metadata
  • AI-generated insights (if analysis is complete)
Source: backend/routes/case.py:146-165

​
Case Status Flow

Cases progress through the following states:
1

Created

Case is initialized with patient information
2

Processing

Documents are being analyzed by AI agents
3

Completed

AI analysis finished, insights available
4

Failed

Error occurred during processing (requires review)

​
File Management

​
Supported File Types

Lab Documents
  • PDF reports
  • Text-based laboratory results
  • Stored with extracted text for analysis
Radiology Files
  • JPEG, PNG medical images
  • DICOM files
  • Processed with vision AI for findings extraction

​
File Storage Structure

Files are organized by category and case: 
{category}_files/{case_id}/{filename}
Example:
  • lab_files/917034a2-c50e-48f4-8289-963ad7b0ad58/blood_test.pdf
  • radiology_files/917034a2-c50e-48f4-8289-963ad7b0ad58/chest_xray.jpg
Source: backend/routes/case.py:68-96

​
Updating Cases

Cases can be updated with new information: 
PUT /cases/cases/{case_id}

Payload:
{
  "patient_name": "Updated Name",
  "case_summary": "Updated clinical notes",
  "status": "completed"
}
Only non-null fields in the update request are modified. Existing data is preserved.
Source: backend/routes/case.py:171-199

​
Background Processing

When a case is created, the system automatically triggers background AI analysis: 
# System automatically calls agentic_process in background
background_tasks.add_task(
    agentic_process,
    case_id=case_id,
    user_id=user_id,
    patient_name=patient_name,
    patient_age=patient_age,
    patient_gender=patient_gender,
    case_summary=case_summary,
    lab_files=lab_files_data,
    radiology_files=radiology_files_data,
)
This ensures:
  • Non-blocking API responses
  • Efficient resource utilization
  • Real-time status updates
Source: backend/routes/case.py:103-113, backend/agentic.py:16-124

​
Data Models

The case management system uses structured data models:

​
CaseInput Model

class CaseInput(BaseModel):
    case_id: str
    patient_data: PatientData
    doctor_case_summary: str
    lab_files: Optional[List[ProcessedFile]] = []
    radiology_files: Optional[List[ProcessedFile]] = []

​
PatientData Model

class PatientData(BaseModel):
    name: str
    age: int
    gender: str
Source: backend/models/data_models.py:8-49

​
Best Practices

  • Use descriptive case summaries for quick identification
  • Tag cases with relevant keywords
  • Regular status checks for pending analyses
  • Ensure PDFs are text-readable (not scanned images)
  • Use high-resolution images for radiology files
  • Upload all relevant documents before creating the case
  • Monitor case status after creation
  • Check failed cases for processing errors
  • Retry with corrected files if analysis fails

​
Security Considerations

  • All case data is associated with a user ID for access control
  • Files are stored in secure cloud storage with authenticated URLs
  • Case retrieval requires proper user authentication
  • Patient data follows healthcare privacy standards

​
Next Steps

Document Processing

Learn how uploaded files are processed

AI Analysis

Understand the AI analysis workflow

Build docs developers (and LLMs) love

Get started for freeTalk to us